The first of these studies " Genetic Variants in Genes Related to Estrogen and Tamoxifen Metabolism and the Risk of Breast Cancer in the NSABP Tamoxifen/Breast Cancer Prevention Trial - was a case/case analysis of 39 SNPs in 19 different genes among 249 women with invasive breast cancer (84 exposed to tamoxifen; 165 placebo). This was a null study, by single SNP association, haplotype and pathway analysis. A manuscript is in preparation. The next project - Genetic Polymorphisms as Risk Factors for Therapy-Related Acute Myeloid Leukemia (t-AML) and Myelodysplasia (MDS) in NCI-Sponsored Cancer Clinical Trials " was initiated in collaboration with Dr. Charles Rabkin, Viral Epidemiology Branch, DCEG. The primary study objective is to investigate genetic variation in drug metabolism and DNA repair genes as potential risk factors for therapy-related acute myeloid leukemia and myelodysplasia (t-AML/MDS). We propose collaborating with the major U.S. clinical trials cooperative groups to conduct a comprehensive investigation of polymorphisms affecting the response to cancer chemotherapy and radiation therapy. Collaborating groups will review their previous and current treatment protocols for patients with incident AML/MDS following treatment for cancer within NCI-sponsored Phase III therapy trials. Controls will be selected from patients on the same protocol and treatment arm, who survived free of AML/MDS at least as long as each case, matched for sex, age, and race. We anticipate accruing 500 case/control sets. We have reviewed cases of AML/MDS referred to NCI"s Cancer Treatment and Evaluation Program between 1995 and 2006; 780 cases were identified. Genomic DNA will be obtained by microdissection of banked tumor biopsies, surgical specimens or from blood samples (where available from more recent studies). The ultimate goal is the identification of genetically-recognizable subsets of patients which might benefit from cancer treatment individually tailored to minimize their risk of harm or maximize their therapeutic benefit. We have conducted a pilot study which demonstrated the ability to generate DNA from formalin-fixed, paraffin-embedded (FFPE) tissue using Rubicon OmniPlex whole genome amplification, with SNP genotyping completion/concordance rates of 99%/99.8% for amplified DNA derived from serum and 97%/99% for amplified DNA extracted from FFPE tissue. Utilizing the resources of the Prostate, Lung, Colon and Ovarian (PLCO) Cancer screening trial, we have been investigating the relationship between the Insulin-Like Growth Factor (IGF) Signaling Pathway and Risk of Advanced Colorectal Adenoma, prompted by data suggesting that IGFs may represent potentially modifiable cancer risk factors. We have analyzed 800 participants found to have an advanced colorectal adenoma at the time of baseline screen, and 800 matched non-adenoma subjects. Genotyping has been completed on 37 SNPs in 7 IGF-related genes (IGF1, IGF-BP3, ALS, IGF-1R, IGF-BP5, IGF2 and GH), and circulating levels of IGF-1, IGF-2 and IGFBP-3 have been measured. The latter documented a 1.7-fold increase in adenoma risk (95% C.I. 1.2-2.5) in highest vs. lowest quartiles of IGF-1, controlled for IGF-2, IGF-BP3 and numerous other covariates. The evaluation of genetic variants as primary risk factors for advance adenoma was null, although we confirmed the previously-observed strong relationship between IGF-BP3-01 (rs2854744), and a new association between IGF-BP3-07 (rs6413441)and circulating levels of IGF-BP3 among controls. These two SNPs decrease IGF-BP-3 levels by 222 and 148 units per minor allele (mean IGF-BP3 level 4,000 units). Manuscripts summarizing these findings are nearing completion. Finally, we are developing a major project aimed at identifying Genetic Modifiers of Osteogenic Sarcoma Risk, building upon pilot observations made by Dr. Savage in the Bone Disease and Injury Study of Osteosarcoma (104 cases, 74 controls). Analysis of 11 growth-related genes identified a functional single nucleotide polymorphism (SNP) in IGF2R that was associated with an odds ratio=2.0; 95% CI 1.3-3.2 (94). Selected TP53 SNPs were associated with 6 to 8-fold increases in osteosarcoma (OS) risk (93), and a protective effect was observed between specific haplotypes in the telomere genes TERF1, POT1 and TEP1 (S. Savage, unpublished data). A collaboration has been established with the Children"s Oncology Group to extend these observations into a series of 900 OS cases and a similar number of controls, using a candidate gene strategy targeting more than 100 genes implicated in growth, height, hormone synthesis and metabolism, bone formation, tumor suppression, cell cycle regulation, chromosome stability, telomere maintenance, DNA repair and ribosomal metabolism.